‘Is Residual Income Model (RIM) REALLY Superior to Dividend Discount Model (DDM)?’ – A Misconception

Abstract

The development of the residual income model (RIM) has potential implications for the empirical researchers as the model specifies relationship between earnings and book values as proxies for equity values and accounting variables. Although researchers have supported RIM as an alternative to the dividend discount model (DDM), some empirical studies on RIM have triggered arguments on the superiority of the RIM over DDM. In theory, both models give the same value estimates; empirically, these value estimates changes with the changes in the assumption sets. In this paper, we show that both models provide the same values estimates when the terminal value can be forecasted. Although, under the perpetual growth rate model, the researchers have shown that empirically RIM outperforms DDM. We have shown that this superiority of RIM is misleading, as the transversality condition, a necessary assumption for deriving the RIM, is void under the perpetual growth rate scenario. Keywords: Book Value, Clean Surplus Relation (CSR), Dividend Discount Model (DDM), Residual Income Model (RIM), Valuation. I. Introduction How do we best measure value creation in companies, and how do we best summarize our expectations about future value creation into an estimate of equity value? Residual Income Model (RIM) and Dividend Discount Model (DDM) are the two most widely used valuation techniques in finance, and in accounting. Researchers have been arguing on the superiority of one model over the other, however. Yet, none of them have investigated the reason behind this difference in the empirical tests of the models. The RIM is an algebraic derivation of the DDM under some robust assumptions. The RIM is based on simplified accounting relationship as well as on the assumption of DDM. The RIM is seductive because it purports to provide assessments of performance at any given point in time. The rejection of RIM is logically equivalent to prices not being equal to the present value of expected future dividends. The researchers deny the fact that the RIM is at fault, rather they have been arguing on the empirical testing methods. Bernard (1995), Penman and Sougiannis (1998), Francis et al. (2000), Frankel and Lee (1998) have argued that RIM provides a better measure of the asset value than that of DDM. Although in theory, both DDM and RIM yield identical value estimates of the intrinsic value; in practice, they will differ if the sets of assumptions differ. In this paper, we focus on the reason behind getting different value estimates from DDM and RIM. We show that the presence of terminal value provides the same value estimates of the models. With the perpetual growth rate, the RIM is regarded as a better measure for valuing an asset. We show that this belief is misleading and both of the models will provide the same value estimate if the transversality condition of RIM holds. The paper is organized as follows. Section II provides a brief discussion on the asset valuation. Section III and IV incorporate residual income model and its development. Arguments over the empirical studies are included in section V. The argument of this paper, the simulation of the RIM under perpetual growth rate is introduced in section VI. The implication of the empirical analysis of this paper is discussed in section VII. Section VIII follows the conclusion. II.